Home

Cover Story

Table of Contents

E-Newsletter

Article Archive

Editorial Calendar

Datebook

Writers' Guidelines

Orgs/Links

Opinion Polls

Reprints

Forum

For other articles and previous issues click here.

August 16, 2004

Pediatric Head Trauma — Looking at the Options
By Jamey West, MD, and Lisa Coronado
Radiology Today
Vol. 5 No. 17 Page 18

Health physicists and other healthcare professionals have concern over the collective radiation dose to the general population, especially children. Doctors commonly order CT scans of the head for children, since often assessment of their mental status and neurological exams are harder to evaluate than in adults.

While there is great clinical benefit gained from the cranial CT scan, there appears to be room for decreasing radiation exposure from it. Just exactly when and to what degree reduction is indicated requires pertinent knowledge about the radiation exposure involved in a CT scan and what that radiation means in terms of risk to the child.

Head trauma is one of the most common situations where cranial CT scans are used in children. In cases of severe head trauma with coma and obvious skull fractures, the need for rapid assessment to save lives makes CT a clear choice. Plain films do not show any detail of intracranial injury. MRI takes much too long and interferes with monitoring of an unstable patient.

The more controversial use of cranial CT scanning in children is for minor head trauma. Thousands of pediatric patients with minor head trauma are seen in emergency departments (EDs) in the United States every year. The vast majority of these children have no intracranial injury, but 1% to 2% will have life-threatening intracranial injuries. Most of these injuries are treated successfully if detected early and repaired by a neurosurgeon, whereas if left to worsen may result in permanent disability or death.

CT scans for head trauma interpreted by competent radiologists have proven extremely reliable for evaluating these cases. At least five studies have shown that a negative CT scan of the head has had zero false negatives—even with skull fracture in many cases. If the CT scan shows no intracranial injury, there is a miniscule chance of neurological deterioration. Because of this wonderful reassurance, most children with minor head trauma are scanned with CT when the technology is available.

Too Much of a Good Thing?
Many people in healthcare believe CT is overutilized, often citing the extremely high number of negative scans. But most screening technologies produce many more negative results than positive ones. To date, cost—not radiation exposure—has been the main drawback to ordering CT scans. However, medicine is learning that relatively low-dose radiation (less than 1 sievert) is associated with an increased incidence of nervous system tumors. A reliable way to differentiate between which pediatric head trauma cases justify CT and which don’t would provide a valuable decision tool for clinicians.

Minor head trauma arguably calls for a CT scan to rule out the small chance of neurological deterioration, but in trivial head trauma cases, the likelihood of damage may be so small that it doesn’t justify the radiation exposure. A practical clinical way to differentiate between the two would reduce the number of children exposed to CT radiation without missing serious intracranial injuries.

One approach to this issue is using the Glasgow Coma Scale (GCS), which is learned by emergency medicine specialists and other clinicians who evaluate head trauma patients. Doctors using the scale score patients from 1 to 5 in each of three categories: eye opening, speech, and movement of limbs. A patient with eyes open, spontaneous appropriate speech, and normal movement of limbs scores 5 in each area for a total score of 15. A comatose patient who will not open his or her eyes, speak, or move limbs despite any stimulation scores 1 in each area for a total score of 3. Minor head trauma patients have GCS of 13 to 15, but trivial head trauma is always 15.

To further differentiate between the two minor trauma types, trivial head trauma is a witnessed head injury that results in no loss of consciousness, amnesia, or any other symptoms. There is no sign of head injury or any neurological deficits on physical examination. The mechanism of injury is not worrisome.

There is an interesting difference in ordering CT scans for head trauma between the United States and various countries around the world. In Italy, head CT is ordered only if there is a skull fracture on plain film. In Denmark, physicians rarely order CT scans of the head—and then only by a neurosurgeon. In the United Kingdom and Spain, head CT is ordered only for skull fracture, neurological deficit, or deteriorating mental status, according to “The Canadian CT Head Rule Study for Patients With Minor Head Injury: Rationale, Objectives, and Methodology for Phase I,” written by Ian G. Stiell, MD, and colleagues and published in the August 2001 issue of Annals of Emergency Medicine. Mandatory training about the radiation dose per radiological study is required in the United Kingdom.

Few Guidelines
One reason U.S. doctors so frequently order CT scans for trauma is that there have not been definite guidelines for emergency physicians. In addition to identifying the clinical combination that is 100% sensitive in ruling out all cases of intracranial injury with minor head trauma, clinicians and hospitals certainly want medical-legal protection as well. In “Evaluation and Management of Children Younger Than Two Years Old With Apparently Minor Head Trauma: Proposed Guidelines,” Sara A. Shutzman, MD, and colleagues divided children under 2 years of age presenting with head trauma into four classes. They published these guidelines in Pediatrics in May 2001.

High risk. These children with a high risk of intracranial injury definitely merit ordering a CT scan. They have classic signs of severe brain injury: depressed mental status, focal neurological deficits, signs of basilar skull fracture, seizures, irritability, acute skull fracture, bulging fontanels, vomiting more than five times in six hours, and loss of conscious for more than one minute.

High-intermediate risk. These patients have some lesser signs of brain injury. An example would be vomiting three to four times, behavioral changes, and nonacute skull fracture. These patients should either be immediately scanned or be observed in the ED for four to six hours.

Intermediate risk. The authors established a second intermediate risk category of patients with concerning history, unknown mechanism of injury, or physical exam with sign of skull fracture. These patients should receive CT, skull films, or be observed in the ED for four to six hours.

Low risk. Lastly, the low risk group may be discharged from the ED with instructions. These patients have normal mental status, normal neurological exams, no loss of consciousness, no vomiting, and no sign of head trauma. The mechanism of injury in this group, which does not need to be scanned, is unequivocally trivial. No patients involved in motor vehicle accidents or with unwitnessed head trauma are placed in this category.

Shutzman and colleagues pointed out that a much lower threshold for cranial CT should occur for patients younger than the age of 1—especially for children younger than 3 to 6 months—because they are harder to assess, have severe intracranial injury with less trauma, and have a higher rate of intentional head trauma. Another reason to obtain CT scans of young infants is that they actually need less medical sedation for the cranial CT than older babies, so sedation is less of a drawback. Medical sedation should be avoided whenever possible because it can cause respiratory depression and aspiration, interfere with monitoring mental status, and even mask deterioration if there’s a life-threatening complication.

Paul Rosman, MD, presented a simple recommendation for a child with minor head trauma. He has shown evidence that it is safe in most cases to omit a CT scan if a child has a GCS of 13 to 15, has suffered no loss of consciousness, and has no focal neurological deficits. That recommendation was included in “Imaging Procedures in Pediatric Neurological Conditions,” published in the October 10, 1998, issue of Pediatric Annals.

If time is available, for instance, in follow-up of a patient with head trauma, MRI is a more sensitive detector of damage to the brain. MRI also gives a better view of the posterior fossa and cerebellum, which has artifact from the dense bone in that region on a CT scan. MRI has the advantage of not exposing the patient to ionizing radiation.

Other CT Uses
Another common use of cranial CT scans in children is in the acute onset of seizures. In the rare instances that CT shows an abnormality, the finding is usually of no clinical significance, according to “The Role of Computed Tomography in Evaluating New Onset Seizures in the Emergency Department,” published in the August 2000 issue of Epilepsia (p. 950-954). History and physical examination are sufficient to identify those patients in whom further studies are necessary. Routine CT scanning is not recommended for patients with no known seizure risk factors, normal neurological exam, no acute symptoms other than fever, and reliable neurological follow-up, according to the article.

One of the most common intracranial problems in premature infants is ventricular hemorrhage. Sonography is the method of choice for detecting the hemorrhage and subsequent pathology, according to Imaging of the Newborn, Infant, and Young Child, written by pediatric radiologist Leonard Swischuk, MD. There are some injuries to newborns that must be evaluated by CT, such as depressed skull fractures that may harbor an underlying subdural hematoma. The ultrasound transducer uses the anterior and posterior fontanels as windows to the brain and cannot be angled well enough to see pathology along the convexities of the brain. If birth trauma is suspected, usually CT must be performed, as it is the preferred method of detecting epidural and subdural hematoma, subarachnoid hemorrhage, and cerebral contusion.

MRI can be a difficult task in newborns because of the difficulty of monitoring the patient. Recently, innovations have been accomplished that allow MRI by shielding special supportive equipment. MRI can show excellent detail, and if not used in the emergent situation, at least can save the infant the many follow-up scans that occur after birth trauma. In the past, it had not been uncommon for an infant to receive four or five serial CT scans for an intracranial injury following birth trauma.

Other experts have said that sonography has occupied the niche for acute problems that can be used at cribside and MRI for the long-term follow-up of disease if more details are needed. CT has been relegated to specific suspicions of certain types of acute hemorrhage and intracerebral calcifications, according to Care of the High-Risk Neonate, by Marshall H. Klaus, MD, and Avroy A. Faranoff, MD. Electroencephalogram is accepted as the method of choice for studying neonatal seizures.

Headaches
Headaches are another patient complaint where hundreds of thousands of head CT scans are performed. The vast majority of those scans are negative or aren’t helpful, according to “Headaches in Children and Adolescents,” written by Donald W. Lewis, MD, and published in the February 15, 2002, issue of American Family Physician. Neuroimaging is not routinely warranted in the evaluation of childhood headache unless there are neurological or behavior changes or the pattern is chronic progressive in nature, according to the same article.

Forty percent of children get headaches by the time they are 7 years of age and 75% complain of headaches to their physician by the age of 15. Reasons to perform CT scans should be limited to excruciating pain, papilledema on eye exam, and multiple episodes of vomiting (signs of increased intracranial pressure), according to “A Guide to Children with Acute and Chronic Headaches,” written by Ruth K. Rosenblum, RN, MS, CPNP, and Paul Graham Fisher, MD, and published in the September/October 2001 issue of the Journal of Pediatric Health Care.

MRI is actually replacing the CT scan as the treatment of choice, showing much more detail. Cost is probably the main factor that influences some physicians today to order a CT scan over an MRI. Hopefully, as more physicians and insurers appreciate the real risk of radiation, even fewer head CT scans will be ordered for headache evaluation.

In a life-threatening emergency, the risk to the patient from radiation exposure is comparatively minor. On the other hand, when considering that hundreds of thousands of patients will be scanned every year, the increased cancer risk to the population must be considered. There may not be a reduction in head CT scans in the United States to the level of that in Europe, but with published guidelines physicians should feel safer—from a clinical as well as medical-legal point of view—in foregoing some CT scans. Use of alternatives should be encouraged whenever feasible, such as MRI and ultrasound. Other innovations, such as an observational period in pediatric EDs, will relieve some of the anxiety of using CT scan to prevent “ever missing anything.”

— Jamey West, MD, is lead technical reviewer for the radiation safety committee for the National Institutes of Health.

— Lisa Coronado is senior health physicist for the National Institutes of Health.

Subscribe to Radiology Today Magazine!